---

base_model:

• infly/Infinity-Parser2-Flash

tags:

• text-generation-inference
• llama-cpp
• ocr
• pdf
• document-parsing
• document-understanding
• layout-analysis
• table-recognition
• chart-parsing
• formula-recognition
• chemical-formula
• markdown
• vision-language
• infinity-parser
• infinity_parser2

license: apache-2.0

language:

• en
• zh

pipeline_tag: image-text-to-text

library_name: transformers

---

Infinity-Parser2-Flash-GGUF

> Infinity-Parser2-Flash is a low-latency document understanding model from infly-ai, one of two variants in the Infinity-Parser2 flagship family (alongside the accuracy-optimized Infinity-Parser2-Pro), engineered for fast inference while consolidating robust multi-modal parsing into a unified architecture trained via an upgraded synthetic data engine spanning nearly 5 million diverse document samples and a novel multi-task reinforcement learning approach with verifiable rewards across document parsing, element parsing, chart parsing, chemical formula parsing, document VQA, and general multimodal understanding. It delivers a 3.68x speedup over the previous Infinity-Parser-7B model (increasing throughput from 441 to 1,624 tokens/sec) while still posting strong benchmark results — 86.0% on olmOCR-Bench, 72.2% on ParseBench, and 91.98% on OmniDocBench-v1.6 — outperforming frontier models like DeepSeek-OCR-2 and MinerU2.5 on several document-parsing tasks, though trailing its larger Pro sibling on layout analysis, chart/chemical formula parsing, and general multimodal benchmarks (e.g., MMMU, AI2D, MathVista). It extracts structured layout with bounding boxes, category labels, and per-element text (LaTeX for formulas, HTML for tables, Markdown for text), supports command-line and Python API usage via the infinity_parser2 package with vLLM, transformers, or vLLM-server backends, and is released under Apache-2.0, with known limitations primarily around English/Chinese-only support, degraded accuracy on complex charts and rotated table elements, and no fine-grained text formatting (bold, italic, strikethrough) capture.

Model Files

File Name | Quant Type | File Size | File Link |

|-----------|------------|-----------|-----------|

| Infinity-Parser2-Flash.BF16.gguf | BF16 | 3.78 GB | Download |

| Infinity-Parser2-Flash.F16.gguf | F16 | 3.78 GB | Download |

| Infinity-Parser2-Flash.F32.gguf | F32 | 7.54 GB | Download |

| Infinity-Parser2-Flash.Q2_K.gguf | Q2_K | 969 MB | Download |

| Infinity-Parser2-Flash.Q3_K_L.gguf | Q3_K_L | 1.16 GB | Download |

| Infinity-Parser2-Flash.Q3_K_M.gguf | Q3_K_M | 1.1 GB | Download |

| Infinity-Parser2-Flash.Q3_K_S.gguf | Q3_K_S | 1.02 GB | Download |

| Infinity-Parser2-Flash.Q4_0.gguf | Q4_0 | 1.2 GB | Download |

| Infinity-Parser2-Flash.Q4_K_M.gguf | Q4_K_M | 1.27 GB | Download |

| Infinity-Parser2-Flash.Q4_K_S.gguf | Q4_K_S | 1.21 GB | Download |

| Infinity-Parser2-Flash.Q5_0.gguf | Q5_0 | 1.37 GB | Download |

| Infinity-Parser2-Flash.Q5_K_M.gguf | Q5_K_M | 1.41 GB | Download |

| Infinity-Parser2-Flash.Q5_K_S.gguf | Q5_K_S | 1.37 GB | Download |

| Infinity-Parser2-Flash.Q6_K.gguf | Q6_K | 1.56 GB | Download |

| Infinity-Parser2-Flash.Q8_0.gguf | Q8_0 | 2.01 GB | Download |

| Infinity-Parser2-Flash.mmproj-bf16.gguf | mmproj-bf16 | 671 MB | Download |

| Infinity-Parser2-Flash.mmproj-f16.gguf | mmproj-f16 | 671 MB | Download |

| Infinity-Parser2-Flash.mmproj-f32.gguf | mmproj-f32 | 1.33 GB | Download |

| Infinity-Parser2-Flash.mmproj-q8_0.gguf | mmproj-q8_0 | 365 MB | Download |

llama.cpp

LLM inference in C/C++ — https://github.com/ggml-org/llama.cpp